| 9fbfce6bd98f1d790f75215932de854892a58154 |
|
24-Nov-2017 |
Wind Yuan <feng.yuan@intel.com> |
headers: add xcam_std.h as xcam cplusplus first header
* new xcam_std.h as first header instead of xcam_utils.h
* all export headers with #include <> instead of #include ""
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6987f2ba7c159a8916a876f2b4d7505a34c97f10 |
|
30-Sep-2017 |
Yinhang Liu <yinhangx.liu@intel.com> |
standard-OCL: remove libdrm dependence
* disable drm as default
* configure cmdline:
$ ./autogen.sh --prefix=/usr --enable-gst --enable-libcl \
--enable-profiling --enable-smartlib --enable-opencv
/external/libxcam/tests/test-cl-image.cpp
|
| cff638b59385036cbc91873a9965282c8861a306 |
|
25-Sep-2017 |
Yinhang Liu <yinhangx.liu@intel.com> |
standard-OCL: switch to CLVideoBuffer path
* CLVideoBuffer inherited from BufferProxy and CLBuffer
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6b5f71042b969ae41362e07660181a676685702b |
|
15-Sep-2017 |
Yinhang Liu <yinhangx.liu@intel.com> |
standard-OCL: remove DrmBoBuffer dependency
* use VideoBuffer instead of DrmBoBuffer
* encapsulate the CLImage/CLBuffer creation function
/external/libxcam/tests/test-cl-image.cpp
|
| 0995e53b9064a563b343a9bdcd50ecf88bf84bec |
|
23-Aug-2017 |
Wind Yuan <feng.yuan@intel.com> |
module: add stitcher module interface
* module/interface, new feature interface entry.
* add interface/stitcher.h as stitching module interface.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6d144ce2e4095c39b7776f308e9dd5bb4792b1b3 |
|
16-Aug-2017 |
Wind Yuan <feng.yuan@intel.com> |
framework: extract generic buffer functions to base
* weaken BufferProxy and enhance VideoBuffer
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 75f79419d0f76d18c23fc3aeca30ee8286c62682 |
|
10-Jul-2017 |
Wind Yuan <feng.yuan@intel.com> |
robust: fix potential mem-leak from klocwork scan
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 4322c5668e7751452098015067f015a3432aeedd |
|
08-Jul-2017 |
Yinhang Liu <yinhangx.liu@intel.com> |
cl-binary: simplify usage of binary kernel
* simplify test-binary-kernel source code
* remove clx_bin directory
* build binary kernel at runtime
* create_cl_binary_demo_image_handler is an example
* test-binary-kernel cmdline:
$ test-binary-kernel --src-kernel kernel_demo.cl \
--bin-kernel kernel_demo.cl.bin --kernel-name kernel_demo
* test-cl-image cmdline:
$ test-cl-image -t demo -f BA10 -i input.raw \
-o output.raw -k kernel_demo.cl.bin
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 7110f89ce125e8d4dedcd585fea7b44636403400 |
|
30-Jun-2017 |
Yinhang Liu <yinhangx.liu@intel.com> |
file-path: standardize the path of header files
* add module names (e.g. ocl/isp) for headers
* correct cflags to "-I${prefix}/include/xcam -I${prefix}/include"
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| be505049e0d0cd218324c728b840652ac54bd19f |
|
01-Jun-2017 |
Wind Yuan <feng.yuan@intel.com> |
cl-framework: refine cl kernel handler arguments
* add cl argument class, template and array for different types
for example.
CLArgList args;
args.push_back (new CLMemArgument (cl_image));
args.push_back (new CLArgumentT<int> (1920));
* cl kernel support async mode (disable block), cl arguments
would be released by event hook.
* CLImageKernel arguments can be set by deriving <prepare_arguments> or
directly <set_arguments>. only one can work.
* CLImageHandler, all parameters should be ready in <prepare_parameters>
and released in <execute_done>
* Image input/output buffer can be found via CLImageHandler
get_input_buf ()
get_output_buf ()
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 5aad612ed06d2011cdc914555101cf5e01bfd31e |
|
19-Mar-2017 |
Yinhang Liu <yinhangx.liu@intel.com> |
cl-binary: move kernel binary to text as built-in
* How to load binary-text file:
1. generate binary file, related script:
libxcam/tests/test-binary-kernel
Usage:
$ test-binary-kernel -i kernel_demo.cl -o kernel_demo.cl.bin \
-k kernel_demo
2. generate binary-text file, related script:
libxcam/tools/convert-binary-to-text.sh
Usage:
$ convert-binary-to-text.sh kernel_demo.cl.bin \
kernel_demo.clx.bin
3. include binary-text file in create-handler function, please refer
to demo handler:
SmartPtr<CLImageHandler>
create_cl_binary_demo_image_handler (SmartPtr<CLContext> &context)
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 9d5e6a3d98eabe4727b9d8a4333ff90ef2540269 |
|
26-Jan-2017 |
Wind Yuan <feng.yuan@intel.com> |
cl-modules: obsolete unused handlers
* delete handlers never used.
cl_blc_handler
cl_wb_handler
cl_bnr_handler
cl_macc_handler
cl_snr_handler
cl_hdr_handler
* mark arguable unused handlers into obsolete without build,
may remove them anytime without notice.
cl_gamma_handler
cl_3a_stats_calculator
cl_dpc_handler
cl_ee_handler
cl_demosaic_handler
* merge cl_denoise_handler and cl_biyuv_handler into
cl_bilateral_handler
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 95b94fe4eb62f3945045cf69216631c156b98deb |
|
05-Dec-2016 |
Wind Yuan <feng.yuan@intel.com> |
test-cl-image: fix fisheye wide angle
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 74f9fdaec5d67618e5c08a6ed9ce61c9f9a3ad0d |
|
18-Nov-2016 |
Wind Yuan <feng.yuan@intel.com> |
cl-fisheye: support fisheye correction
* map fisheye to equirectangular coordinate(GPS position)
* test cmdline:
$ ./test-cl-image -i fisheye_1080p.nv12 -t fisheye \
-o gps_location.nv12 -f NV12 -g NV12
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 213b5d689ee8bd440bddf12f20540670c4bb3fd4 |
|
25-Nov-2016 |
zongwave <wei.zong@intel.com> |
cl-framework: fix cl processor buffer queue block issue
* fix disordering issues from check_ready_buffers
* fix gstreamer buf-block issue in xcamfilter
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 1f31ef0ea04242f581d1c37f21b47cff931c93b4 |
|
08-Nov-2016 |
Wind Yuan <feng.yuan@intel.com> |
cl-handler: fix by-pass value check
* CLImageHandler::execute may return XCAM_RETURN_BYPASS if
<prepare_parameters> is not ready
* test-cl-image continues on next-buffer processing when <execute> return
XCAM_RETURN_BYPASS
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 233fe1a70a00dd094cfa8fe58d5ebdbe91c2e72a |
|
04-Nov-2016 |
zongwave <wei.zong@intel.com> |
cl-warp: add image warp test case in test-cl-image
* test commandline:
$ ./test-cl-image -t warp -f NV12 -g NV12 -i TV_1080P_NV12.yuv
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| e676aec5b2125467eb78e471fc0b3e0389cbf297 |
|
27-Sep-2016 |
Yinhang Liu <yinhangx.liu@intel.com> |
tests: fix read buffer failed issue in test-cl-image
* the while loop should break at the end of file
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 136aa4f5809c82471d07e68270bd3c8aa5c9a533 |
|
08-Sep-2016 |
Wind Yuan <feng.yuan@intel.com> |
robust: fix klocwork issue and convenient image write
* fix klocwork issue of copy construction
* replace image-read/write to ImageFileHanle to make it
convenient in tests
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| b43f99c0ca903a62838628dc1c06f3d5fb576746 |
|
22-Aug-2016 |
zongwave <wei.zong@intel.com> |
cl-3d-denoise: use sub-group to improve performance
* 2 versions supported
- 1st, use both sub-group and SLM to reach better performance
- 2nd, pure SLM version can work cross platforms and
with good performance.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| bec89e12eafc649573ed9fa771a2ac1ea4bb54aa |
|
22-Jul-2016 |
zongwave <wei.zong@intel.com> |
cl-3d-denoise: enable 3D denoise CL kernel/handler
* test cmdline:
$ ./test-cl-image -t 3d-denoise -f NV12 -g NV12 \
-i noisy_sequence_nv12.yuv -o 3d_denoise.yuv
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 98d23209aed256b292214659d89112dccbe75ebc |
|
27-Jun-2016 |
zongwave <wei.zong@intel.com> |
cl-wavelet-nr: add interface to enable bayes shrink
* wavelet-mode specify wavelet denoise mode, default is off select from
[0:disable, 1:Hat Y, 2:Hat UV, 3:Haar Y, 4:Haar UV, 5:Haar YUV,
6:Haar Bayes Shrink]
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| cc33d31edd6b0b00e785ba97d0b1af7dd28a53bc |
|
07-Jun-2016 |
zongwave <wei.zong@intel.com> |
cl-wavelet: improve performance for bayes shrink
* use share local memory and mad operator to improve performance
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 8109f14bd88ee35c34cf47758dbab83518dada46 |
|
25-May-2016 |
zongwave <wei.zong@intel.com> |
cl-wavelet: optimize estimation of bayes threshold
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| f4e17ab3b9c58451ee7a9e67ead0ad6c5a2c5560 |
|
04-May-2016 |
Wind Yuan <feng.yuan@intel.com> |
cl-defog: feature supported in tests and plugin
* test-cl-image:
$ ./test-cl-image -t dcp -f NV12 -g NV12 -i fog.nv12 -o defog.nv12
* test-device-manager for dcp
$ ./test-device-manager -m dma -c -f BA10 -d still -p -a dynamic \
--pipeline=advance --defog-mode dcp
* test-device-manager for retinex
$ ./test-device-manager -m dma -c -f BA10 -d still -p -a dynamic \
--pipeline=advance --defog-mode retinex
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6b3e260deeb8ee6f157dd11c38f9e75ab094baef |
|
25-Apr-2016 |
Yinhang Liu <yinhangx.liu@intel.com> |
tests: add psnr calculation in test-cl-image
* support PSNR_Y, PSNR_R, PSNR_G, PSNR_B
* test command line:
# test-cl-image -t [type] -f NV12 -r refer.nv12 \
-i input.nv12 -o output.nv12 -P
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 8b3bf7a898e54a3b7fa4c7825ac99a9d795b83ff |
|
01-Apr-2016 |
zongwave <wei.zong@intel.com> |
cl-wavelet: support image on any resolution
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| fde746ddc36cd76394347aad055f2d0a168379ad |
|
30-Mar-2016 |
zongwave <wei.zong@intel.com> |
cl-wavelet: tune threshold of Haar wavelet denoise
* Haar-wavelet shrinkage denoise threshold was tuned to improve
quality in ultra low light.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 7db344d95e2d3025509bfa9b7f3eecbdecce5a6e |
|
30-Mar-2016 |
zongwave <wei.zong@intel.com> |
tests: fix swap buffer order in test-cl-image
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| b7140441fcd8da54d05ee5598c720535380d4f16 |
|
25-Mar-2016 |
zongwave <wei.zong@intel.com> |
cl-wavelet: support Haar wavelet in both Y and UV channels
* support Haar-wavelet decomposition and reconstruction in both
Y and UV channels
* fix test-cl-image issues on swapped-buffer
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 0ad35acf97e48002078c1cc3100c1169e46d71c4 |
|
28-Mar-2016 |
Wind Yuan <feng.yuan@intel.com> |
robust: fix potential issues reported by klocwork
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| bc1e07124ab6930b3462042eb4977f00b94c5a8d |
|
01-Mar-2016 |
Wind Yuan <feng.yuan@intel.com> |
tests: add wavelet test case
* test cmdline:
$ ./test-cl-image -t wavelet -f NV12 -g NV12 -i input_1080p.nv12 \
-o out_1080p.nv12
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6523286e77ad1338a3362fe1f6b7a1c115168ecf |
|
25-Feb-2016 |
Wangfei <feix.w.wang@intel.com> |
cl-gauss: add gaussian-blur handler on Y format.
* test cmdline:
./test-cl-image -t gauss -f NV12 -i in.nv12 -o out.nv12
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 8e5e3e4bb0c6f1fd26b577d3cc21020fb2b338c6 |
|
18-Feb-2016 |
Wangfei <feix.w.wang@intel.com> |
cl: add retinex algorithm to enhance edge on yuv
* test cmd:
./test/test-device-manager -f BA10 -m dma -c -d still -p -a \
dynamic --enable-retinex
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| cd70194563f95458bf5ec917c0a292b2498fb621 |
|
28-Oct-2015 |
Wind Yuan <feng.yuan@intel.com> |
format: support RGB planar format
* support new format XCAM_PIX_FMT_RGB48_planar and
XCAM_PIX_FMT_RGB24_planar
* fix get_planar_info argument
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 43135774c9809eef3d6645cbcc008b36d1cf8201 |
|
09-Sep-2015 |
Wind Yuan <feng.yuan@intel.com> |
test: add bayer-nr option for cl-test-image
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 9845b02a95b2219e684c8b0d5ece06fe4fb7ad84 |
|
27-Aug-2015 |
wujunkai166 <wujunkai166@gmail.com> |
tests: fix format from rg12 to ba12 in tests
* test-device-manager: Change rg12 to ba12
* test-cl-image: Change rg12 to ba12 and change format to
V4L2_PIX_FMT_SGRBG12.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| bf15dda28e073b282e9ae71af9307d35545cf2eb |
|
19-Aug-2015 |
Yinhang Liu <yinhangx.liu@intel.com> |
tests: fix NV12 read/write failed on test-cl-image
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6e05115ccae00564c2fad6e02d16230ca38459d5 |
|
14-Aug-2015 |
yaowang1 <yao.y.wang@intel.com> |
cl: add tonemapping framwork on rgba process
* Details:
for tonemapping pipe/kernel:
input: RGBA16
output: RGBA8
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 25787e873efe562d7308f5a3d1963bb1030cebe8 |
|
11-Aug-2015 |
Yinhang Liu <yinhangx.liu@intel.com> |
test-cl-image: fix stride issue
/external/libxcam/tests/test-cl-image.cpp
|
| 347f7043d4203022ce1ac8653de70cc40f2f340c |
|
29-Jul-2015 |
Wangfei <feix.w.wang@intel.com> |
cl pipe: merge all yuv kernel together to improve performance.
* merged macc, csc(rgbatonv12) together.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 1c90c3e61f731e076d8cdd3d3e6825f93457a60d |
|
27-Jul-2015 |
Wind Yuan <feng.yuan@intel.com> |
cl test: add bayer-pipe kernel
/external/libxcam/tests/test-cl-image.cpp
|
| 63e2c26554d5e9f2d279e5c0afca32f467b07982 |
|
26-Jun-2015 |
ShincyTu <shincy.tu@intel.com> |
cl: enable bayer noise reduction feature in testcase
* both test-cl-image and test-device-manager,
command is:
./test-cl-image -t bnr -f BA10 -i input.rggb -o output.rggb
./test-device-manager -f BA10 -m dma -c -d still -p -a dynamic --enable-bnr
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 55bef7d103c2f107d7c59301f8bc59473181df8e |
|
26-Jun-2015 |
ShincyTu <shincy.tu@intel.com> |
cl: enable Defect Pixel Correction for CL path
command is:
./test-cl-image -t defect -f ba10 -i capture_buffer.raw -o dpc.raw
/external/libxcam/tests/test-cl-image.cpp
|
| 2c0a48fb252211c6e5e8c25d663849afc25b5a82 |
|
16-Jun-2015 |
wangfei <feix.w.wang@intel.com> |
add edge enhancement function.
* test ee cmd:
$./test-cl-image -t ee -f NV12 -i capture.nv12 -o ee.nv12 .
/external/libxcam/tests/test-cl-image.cpp
|
| 7f509fa94ccd602ccd8d1008e72e59a2be129be0 |
|
11-Jun-2015 |
ShincyTu <shincy.tu@intel.com> |
cl: add format rg12 in test-cl-image to enable test-cl-image unit test for build-in WDR
/external/libxcam/tests/test-cl-image.cpp
|
| 8df40e3e8d0df4957a611e3a7676ca3a3209d950 |
|
27-May-2015 |
wangfei <feix.w.wang@intel.com> |
csc module: add nv12 to rgba function.
/external/libxcam/tests/test-cl-image.cpp
|
| 35601bd44d082de6553d149aaa3f1acc0edec660 |
|
06-May-2015 |
wangfei <feix.w.wang@intel.com> |
make matrix can be set up when use it in rgba to nv12 conversion.
/external/libxcam/tests/test-cl-image.cpp
|
| 42d62b1bccdc1a46b06b1b80b24c6af7036fb65b |
|
30-Apr-2015 |
wangfei <feix.w.wang@intel.com> |
add macc function.
* test macc cmd:
$./test-cl-image -t macc -f rgba -i capture.rgba -o macc.rgba
/external/libxcam/tests/test-cl-image.cpp
|
| eace4f2ed9fbe22112fc691f95cf912cc8b1df80 |
|
29-Apr-2015 |
ShincyTu <shincy.tu@intel.com> |
cl: add simple noise reduction
* add snr in both test-cl-image and local preview,
the noise reduction algorithm is average filter.
* test command line:
./test-cl-image -t snr -f rgba -i demosaic.rgba -o snr.rgba
or
./test-device-manager -f BA10 -m dma -c -d still -p
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 17dd4ff178dc42fbee4305fe037d47d2ae0bb281 |
|
14-Apr-2015 |
ShincyTu <shincy.tu@intel.com> |
cl: add configuration setting interface for black level correction
/external/libxcam/tests/test-cl-image.cpp
|
| 227ba2742550ba2e800ca45ca723bc1f87766425 |
|
16-Apr-2015 |
wangfei <feix.w.wang@intel.com> |
refine test-cl-image.
* use -g option to specify output format.
* add rgba64 to rgba32 in csc module.
* delete unuseless file kernel_csc.cl.
/external/libxcam/tests/test-cl-image.cpp
|
| 7e0436ec3835e76dc94645f4f7796fbf1e194d0d |
|
15-Apr-2015 |
wangfei <feix.w.wang@intel.com> |
add gamma function.
test gamma cmd:
$./test-cl-image -t gamma -f rgba -i capture.rgba -o gamma.rgba
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| ab5e6fe8bebbc5c766057917d0b2bf590ae2a8f2 |
|
08-Apr-2015 |
Juan Zhao <juan.j.zhao@intel.com> |
add bi-laterial filter for denoise feature
bi-laterial was an algorithm published on Proceedings of the 1998 IEEE
International Conference on Computer Vision, Bombay,India
The paper name is "Bilateral Filtering for Gray and Color Images"
It can smooths images while preserving edges, by means of a nonlinear
combination of nearby image values. It was broadly used in camera for the
features like "buity your skin" or "denoise".
This patch provided an CL implementation with 5x5 window size.
Test Usage: ./test-cl-image -f RGBA -t denoise -i <inputfile> -o <outputfiile>
Signed-off-by: Juan Zhao <juan.j.zhao@intel.com>
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 952dff43dbc618213480b9b4c2e27903c31bf228 |
|
13-Apr-2015 |
Wind Yuan <feng.yuan@intel.com> |
test: fix video buffer init and correct format
* test-cl-image, buffer init doesn't need alignment and
rgba should be format 'V4L2_PIX_FMT_RGBA32'
* cl_demo: image_info.width should be aligned with stride
and image channel bytes (to support rgba format copy)
/external/libxcam/tests/test-cl-image.cpp
|
| 6c16e9844d90b96c79e191abb8118e96d02cb247 |
|
08-Apr-2015 |
Wind Yuan <feng.yuan@intel.com> |
test: change demosaic output format to BGR32
* cl beignet doesn't support RGB32
/external/libxcam/tests/test-cl-image.cpp
|
| 4bbe4ecab625be7f5f2e56d1e87f326dd61a2bf1 |
|
02-Apr-2015 |
Wind Yuan <feng.yuan@intel.com> |
bufferpool: init buffer pool and proxy
* bufferpool, manage all buffer data.
* bufferproxy, protect buffer data from directly access, and
release buffer data to poll when proxy destructed.
* refactor drmbobuffer to derive from bufferpool/proxy
/external/libxcam/tests/test-cl-image.cpp
|
| 125221b9727cffe7ad0b230fec257520179ebc1c |
|
08-Apr-2015 |
wangfei <feix.w.wang@intel.com> |
cl: add hdr process on lab.
test lab hdr cmd:
$./test-cl-image -t hdr -d lab -f rgba -i capture.rgba -o hdr_lab.rgba.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 9842835437b7ef553fa9bb4ed65f45131d895ca4 |
|
02-Apr-2015 |
wangfei <feix.w.wang@intel.com> |
add white balance function.
test white balance cmd:
$./test-cl-image -t wb -f ba10 -i capture.rggb -o whitebalance.rggb
/external/libxcam/tests/test-cl-image.cpp
|
| a1a5ff4a496d784bf43f323c6e6228079605cc1e |
|
30-Mar-2015 |
wangfei <feix.w.wang@intel.com> |
refine csc module.
make _output_format can only change follow with _csc_type.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 1a7d9f3b0e4151dbeae336ee54d97dad8dfc3873 |
|
26-Mar-2015 |
wangfei <feix.w.wang@intel.com> |
add color conversion from rgba to Lab.
cmdline for tests to convert RGBA to Lab:
$ ./test-cl-image -t csc -f rgba -i capture_buffer.rgba -o capture_buffer.lab -c rgba2lab
/external/libxcam/tests/test-cl-image.cpp
|
| 0a1c73e0277787d14222e5509dbd8dcd877ea550 |
|
26-Mar-2015 |
Wind Yuan <feng.yuan@intel.com> |
cl-test-image: standard video_buf replace manual set
/external/libxcam/tests/test-cl-image.cpp
|
| de4c6363a3b06bf311029bf076eb4bf4b60b1fdc |
|
25-Mar-2015 |
Wind Yuan <feng.yuan@intel.com> |
cl csc: remove unnecessary func and hardcode
* no need to have set_output_format since handler name with nv12
* cl kernel doesn't need 1080 hardcode, add vertical_offset for Y
and UV offset
/external/libxcam/tests/test-cl-image.cpp
|
| 5710f97b57db3065e68db063dad59f53f19fe36a |
|
23-Mar-2015 |
ShincyTu <shincy.tu@intel.com> |
cl bayer: enable normalization and black level correction
* Shift 10bit bayer to 16bits to normalize image process
* Support black level correction on default value
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 6ea2221881fa41cbbe9988dedd4917be5c64cc08 |
|
23-Mar-2015 |
wangfei <feix.w.wang@intel.com> |
tests: add color conversion test in cl-test-image.
cmdline for tests to convert RGBA to NV12
$ ./test-cl-image -t csc -f rgba -i capture_buffer.rgba -o capture_buffer.nv12
/external/libxcam/tests/test-cl-image.cpp
|
| 5c96d9ff39a3955c6408d5e2f4d46c5ff701352a |
|
18-Mar-2015 |
Wind Yuan <feng.yuan@intel.com> |
tests: add demosaic test in test-cl-image
cmdline for tests to convert GRBG to RGBA
$ test-cl-image -f BA10 -t demosaic -i capture_buffer.grbg -o capture_buffer.rgba
/external/libxcam/tests/test-cl-image.cpp
|
| cf536697155a16e35f17199ca7a31645c61d9802 |
|
19-Mar-2015 |
wangfei <feix.w.wang@intel.com> |
add print kernel performance function.
usage:
./test-cl-image -t demo -f nv12 -i input.nv12 -o output.nv12 -p 1000
After this it will run demo cl kernel 1000 times and print performance
info and terminate process. If you don't want to use this function,
just don't write '-p 1000' in your cmdline.
Signed-off-by: Wind Yuan <feng.yuan@intel.com>
/external/libxcam/tests/test-cl-image.cpp
|
| 119874612418b92aebb01dc20fca2d7fbc3185d8 |
|
12-Mar-2015 |
wangfei <feix.w.wang@intel.com> |
add HDR function on bayer data.
Currently, this function can deal with bayer data, after proccessing
new bayer data will be put out.
cmdline:
./test-cl-imamge -t hdr -f ba10 -i inputfile -o outputfile.
/external/libxcam/tests/test-cl-image.cpp
|
| 20f67cbe646347772cb1da56a58801aa623b24fe |
|
05-Mar-2015 |
Wind Yuan <feng.yuan@intel.com> |
tests: add cl image test
support CLImageHandler unit tests
cmdline:
./test-cl-image -t demo -f nv12 -i input.nv12 -o output.nv12
/external/libxcam/tests/test-cl-image.cpp
|